owner stringclasses 14
values | repo stringclasses 14
values | id int64 116k 4.61B | issue_number int32 1 181k | author stringlengths 1 39 | body stringlengths 1 262k | created_at timestamp[us]date 2000-06-06 02:40:44 2026-06-02 20:16:18 | updated_at timestamp[us]date 2000-06-06 02:40:44 2026-06-02 20:22:56 | reactions unknown | author_association stringclasses 5
values |
|---|---|---|---|---|---|---|---|---|---|
ClickHouse | ClickHouse | 454,017,307 | 4,050 | den-crane | > +1. I have batches. And it worked ok on MergeTree table. But when I decided to use ReplacingMergeTree no merges are done. Even if I stop INSERTs at all and wait for a day...
> P. s. I do single insert in 10 seconds. But it has small amount of data (less than 1 mb on disc, according to system.parts). I suppose that m... | 2019-01-14T14:10:43 | 2019-01-14T14:10:43 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 455,064,803 | 4,049 | crrlcx | Looks like my issue is solved.
---
The problem was in the __`replicated_can_become_leader`__ parameter. I set it to _`true`_, according to the documentation and description in [MergeTreeSettings.h](https://github.com/yandex/ClickHouse/blob/v18.16.1-stable/dbms/src/Storages/MergeTree/MergeTreeSettings.h).
However,... | 2019-01-17T07:06:26 | 2019-01-17T07:12:30 | {} | NONE |
ClickHouse | ClickHouse | 455,202,756 | 4,074 | KochetovNicolai | There is no restrictions on LowCardinality usage. More information you can find in [this presentation](https://github.com/yandex/clickhouse-presentations/blob/master/meetup19/string_optimization.pdf). | 2019-01-17T15:02:28 | 2019-01-17T15:02:28 | {} | MEMBER |
ClickHouse | ClickHouse | 455,215,547 | 4,063 | alexey-milovidov | How did you generated Cap'n'Proto test message? | 2019-01-17T15:36:23 | 2019-01-17T15:36:23 | {} | MEMBER |
ClickHouse | ClickHouse | 453,921,091 | 4,050 | blinkov | Then that's the reason for the error you see. | 2019-01-14T07:42:28 | 2019-01-14T07:42:28 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 453,971,239 | 4,048 | filimonov | #3276 | 2019-01-14T11:16:02 | 2019-01-14T11:16:02 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 454,009,626 | 3,976 | den-crane | Seems <compiled_expression_cache_size>500 fixes the issue
<img width="953" alt="screen shot 2019-01-14 at 9 42 14 am" src="https://user-images.githubusercontent.com/19737682/51116336-36718300-17e1-11e9-8ac1-b8f006cfe384.png">
metric | value
-- | --
CompiledExpressionCacheBytes | 976000
CompiledExpressionCa... | 2019-01-14T13:46:38 | 2019-01-14T13:46:38 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 454,168,277 | 4,052 | alexey-milovidov | Let's show some compression and performance test. | 2019-01-14T21:28:57 | 2019-01-14T21:28:57 | {} | MEMBER |
ClickHouse | ClickHouse | 454,709,711 | 4,004 | LeoErcolanelli | @filimonov Thanks a lot for the insights, the filtering of the states before feeding them to the backtracking algorithm looks like a great idea!
At the moment I implemented the DFA-based algorithm for the case where there are no `(?t)` in the pattern. I ran a (very) naive test in order to get a feel for the improvem... | 2019-01-16T09:24:50 | 2019-01-16T09:24:50 | {
"+1": 1,
"hooray": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 454,752,124 | 4,049 | crrlcx | > Seems your zookeeper servers not in one cluster.
> Please show
> `echo stat | nc <zookeeper ip> 2181`
> for all zoo servers
* zoo only
```shell
Zookeeper version: 3.4.10-3--1, built on Sat, 03 Feb 2018 14:58:02 -0800
Clients:
/10.130.0.1:44908[0](queued=0,recved=1,sent=0)
/10.130.0.1:56854[1](queued=0,re... | 2019-01-16T11:52:06 | 2019-01-16T11:52:06 | {} | NONE |
ClickHouse | ClickHouse | 455,044,122 | 4,071 | den-crane | Please provide stack-trace(s) from /var/log/clickhouse-server.log | 2019-01-17T05:07:12 | 2019-01-17T05:07:12 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 455,189,310 | 4,074 | alexey-milovidov | > What are the differences from ENUM types?
LowCardinality data type builds dictionaries automatically. It can use multiple different dictionaries if neccessarily.
> Can you give a brief explanation on how it supplements column compression?
It works independent of (generic) columns compression. Columns with Lo... | 2019-01-17T14:25:05 | 2019-01-17T14:28:47 | {
"+1": 3
} | MEMBER |
ClickHouse | ClickHouse | 453,954,546 | 4,046 | filimonov | What kind of client do you use? | 2019-01-14T10:12:07 | 2019-01-14T10:12:07 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 453,955,872 | 4,047 | filimonov | Most probably duplicate of #3891. The issue was fixed in master, but not released yet. | 2019-01-14T10:16:57 | 2019-01-14T10:16:57 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 454,016,189 | 4,050 | den-crane | > ReplicatedMergeTree can not work even 70 items/s.
Zoo MUST be at different physical server. CH utilitizes all 200% I/O and causes Zoo I/O starvation.
Try to install zookeeper on dedicated instance with SSD disks.
And batched inserts. Every insert into replicated table causes Zoo transaction. Zoo transaction wait... | 2019-01-14T14:07:10 | 2019-01-14T14:29:48 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 454,344,465 | 4,026 | wawanawna | > > No effect, so cause is somewhere else
>
> Just to be sure - can you try to remove that condition? Anyway you "insert only last minute data".
>
> Why: maybe now() is beeing calculated once - at the moment of MV initialization?
Hello, thx for suggestion, I've tried create MV without any conditions, didn't he... | 2019-01-15T10:33:44 | 2019-01-15T10:33:44 | {} | NONE |
ClickHouse | ClickHouse | 454,327,005 | 4,053 | danlark1 | Some details i should mention.
MultiVolnitsky saves into the hashtable the index of a needle and a position of it. Then the simple version is used.
3 functions were added
**multiPosition(haystack, [needle_1, ..., needle_n])** -- finding all the positions of all needles, if string was not found, 0 is returned. Re... | 2019-01-15T09:38:05 | 2019-01-15T17:26:44 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 454,710,163 | 4,004 | LeoErcolanelli | Should I go and implement the approach with the discrimination of the states by the DFA before feeding them to the backtracking algorithm (for the patterns containing `(?t)`) or should this PR be merged first so we keep the set of changes small ? | 2019-01-16T09:26:07 | 2019-01-16T09:26:07 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 454,941,597 | 4,078 | alesapin | if all tests are ok then merge | 2019-01-16T21:06:53 | 2019-01-16T21:06:53 | {} | MEMBER |
ClickHouse | ClickHouse | 455,206,247 | 4,063 | alexey-milovidov | I have some troubles reading this code and I've removed the function `checkEqualFrom` by mistake. | 2019-01-17T15:12:07 | 2019-01-17T15:12:07 | {} | MEMBER |
ClickHouse | ClickHouse | 455,457,291 | 4,086 | crrlcx | Hi!
What allocator you use?
I did try jemalloc instead linux default and it works with 128GB RAM and max_memory_usage=60000000000 | 2019-01-18T07:52:32 | 2019-01-18T07:52:32 | {} | NONE |
ClickHouse | ClickHouse | 453,923,753 | 4,049 | M1ha-Shvn | The same result is seen, when I try to drop a partition with:
`ALTER TABLE tbl_test DROP PARTITION 201901;` | 2019-01-14T07:59:03 | 2019-01-14T07:59:03 | {} | NONE |
ClickHouse | ClickHouse | 453,967,868 | 4,046 | eagle1981 | yandex/graphouse :latest | 2019-01-14T11:02:47 | 2019-01-14T11:02:47 | {} | NONE |
ClickHouse | ClickHouse | 454,242,849 | 4,051 | yanglfyangl1978 | We didn't compiled by ourself, but use the release version.
And also, it not happened always, when we reinstall the server, it works. it happened like 1/10 chance as we tested.
Next are the info for binary and platform.
> Description: Ubuntu 18.04.1 LTS
> Release: 18.04
> Codename: bionic
... | 2019-01-15T02:18:17 | 2019-01-15T02:18:17 | {} | NONE |
ClickHouse | ClickHouse | 454,030,198 | 3,669 | Enmk | Performance comparison result.
Performance was measured on my machine against base version (without these modifications) using performance tests included in the PR. Negative numbers represent degraded performance, positive - improved, see `compare.py` for details.
Please note that I've observed +/-1% variance of ... | 2019-01-14T14:49:31 | 2019-01-15T04:35:29 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 454,637,012 | 4,061 | nemoaa | is not working! | 2019-01-16T03:19:10 | 2019-01-16T03:19:10 | {} | NONE |
ClickHouse | ClickHouse | 454,753,637 | 4,049 | crrlcx | Replicated tables are are correctly updated with zookeeper via http ch-to-ch connection over tcp/9009 port.
PS: I also updated the issue topic. | 2019-01-16T11:57:47 | 2019-01-16T17:10:12 | {} | NONE |
ClickHouse | ClickHouse | 455,287,171 | 4,063 | alexey-milovidov | Ok.
The code is a bit difficult to understand for me. I have rewritten it a little and now the test passed.
But I'm not sure that my modifications are correct. Can you please review my modifications?
(and add more tests if needed in separate PR) | 2019-01-17T18:53:55 | 2019-01-17T18:54:28 | {} | MEMBER |
ClickHouse | ClickHouse | 453,931,136 | 4,046 | alesapin | Which version do you use? | 2019-01-14T08:38:25 | 2019-01-14T08:38:25 | {} | MEMBER |
ClickHouse | ClickHouse | 454,246,882 | 4,046 | eagle1981 | Yea, sure.
Working configuration cut-off:
```
#Clickhouse
graphouse.clickhouse.host=10.99.127.130
graphouse.clickhouse.hosts=10.99.127.130
graphouse.clickhouse.port=8123
graphouse.clickhouse.db=graphite
graphouse.clickhouse.compress=false
graphouse.clickhouse.user=graphouse
graphouse.clickhouse.password=*... | 2019-01-15T02:40:06 | 2019-01-15T02:40:06 | {} | NONE |
ClickHouse | ClickHouse | 454,689,533 | 4,070 | blinkov | I'm not aware of any alternative similar tools. You could start with some more simple table and gradually come closer to real one to figure out what exactly is not working. Or just manually convert your data to one of formats supported by ClickHouse https://clickhouse.yandex/docs/en/interfaces/formats/
Link to your ... | 2019-01-16T08:13:50 | 2019-01-16T08:13:50 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 454,003,941 | 4,027 | baldurgaldur | I ran the SELECT query with `final` keyword after the table name:
```SQL
select
sum(tickets) as tickets
from(
select
sum(item_count) as tickets,
max(tracked_to) as tracked_to
FROM transactions FINAL
GROUP BY transaction
HAVING tracked_to = '12'
)
```
Yo... | 2019-01-14T13:26:59 | 2019-01-14T13:26:59 | {} | NONE |
ClickHouse | ClickHouse | 454,161,473 | 4,004 | alexey-milovidov | > Yep, looks like context is not passed to Aggregate functions (while being available in regular functions). Maybe it was done like that by intent? @alexey-milovidov ?
No, it just was not needed yet. | 2019-01-14T21:06:51 | 2019-01-14T21:07:41 | {} | MEMBER |
ClickHouse | ClickHouse | 454,688,220 | 4,070 | alex-zaitsev | Altinity replicator works in production in some companies, so it is reliable enough. Thanks for the problem report, we will look what's wrong.
Continuent recently released replication to ClickHouse as well, you can try it out: https://www.continuent.com/replicating-data-into-clickhouse/ | 2019-01-16T08:08:24 | 2019-01-16T08:08:24 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 454,852,234 | 4,004 | alexey-milovidov | Can you also use `dfa_match` if possible for `sequenceCount`? | 2019-01-16T16:48:53 | 2019-01-16T16:48:53 | {} | MEMBER |
ClickHouse | ClickHouse | 455,417,642 | 2,154 | zzn01 | same issue here, any update? | 2019-01-18T03:42:02 | 2019-01-18T03:42:02 | {} | NONE |
ClickHouse | ClickHouse | 453,975,193 | 4,046 | filimonov | As I understand graphouse uses official JDBC driver.
AFAIK official JDBC diver uses HTTP protocol to connect to ClickHouse server,
Looking on the stacktrace it's looks like the client was connected via TCP protocol.
Maybe you're connecting to wrong port?
For HTTP it's 8123 by default, not 9000. | 2019-01-14T11:31:41 | 2019-01-14T11:31:41 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 454,169,561 | 4,055 | alexey-milovidov | Does not build (`-Wshadow -Werror`): variables declared in `if` are visible in `else`. | 2019-01-14T21:33:06 | 2019-01-14T21:33:06 | {} | MEMBER |
ClickHouse | ClickHouse | 454,702,881 | 4,072 | alesapin | Need for test run -- **don't merge** | 2019-01-16T09:01:36 | 2019-01-16T09:01:58 | {} | MEMBER |
ClickHouse | ClickHouse | 455,041,694 | 4,071 | ggservice007 | A lot of HTTPHandlers | 2019-01-17T04:51:08 | 2019-01-17T04:51:08 | {} | NONE |
ClickHouse | ClickHouse | 455,131,779 | 4,063 | Miniwoffer | Okay, not sure what you want me to use instead of the "checkEqualFrom" function, please elaborate. | 2019-01-17T11:01:13 | 2019-01-17T11:01:13 | {} | NONE |
ClickHouse | ClickHouse | 455,278,463 | 4,063 | Miniwoffer | With the cli tool
`capnp convert json:binary test.capnp CapnProto < test.json`
I can supply the json file if needed | 2019-01-17T18:30:19 | 2019-01-17T18:30:19 | {} | NONE |
ClickHouse | ClickHouse | 453,949,018 | 4,050 | M1ha-Shvn | +1. I have batches. And it worked ok on MergeTree table. But when I decided to use ReplacingMergeTree no merges are done. Even if I stop INSERTs at all and wait for a day...
P. s. I do single insert in 10 seconds. But it has small amount of data (less than 1 mb on disc, according to system.parts). I suppose that may c... | 2019-01-14T09:51:29 | 2019-01-14T09:53:39 | {} | NONE |
ClickHouse | ClickHouse | 454,019,295 | 4,049 | den-crane | Seems your zookeeper servers not in one cluster.
Please show
`echo stat | nc <zookeeper ip> 2181`
for all zoo servers | 2019-01-14T14:17:03 | 2019-01-14T14:17:17 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 454,734,460 | 3,987 | yurigazinskiy | I think its related with MergeTreePartition::getID function [source code](https://github.com/yandex/ClickHouse/blob/cec49357da502c511e006ea8b5d8fbffeb521478/dbms/src/Storages/MergeTree/MergeTreePartition.cpp)
Behaves different for mergetree and replicatedmergetree. | 2019-01-16T10:44:56 | 2019-01-16T10:44:56 | {} | NONE |
ClickHouse | ClickHouse | 454,874,693 | 4,053 | danlark1 | After some optimizations, benchmarks are like this:
```
{
"min_time": 0.589000,
"query": "select count(position(URL, 'yandex')), count(position(URL, 'google')) FROM hits_100m_single"
},
{
"min_time": 0.599000,
"query": "select count(multiPosition(URL, ['yandex', 'google'])) FROM hits_100m_sing... | 2019-01-16T17:50:43 | 2019-01-16T17:50:43 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 455,348,315 | 4,068 | filimonov | Similar #3236 https://github.com/yandex/ClickHouse/issues/1761#issuecomment-449999031 | 2019-01-17T22:01:40 | 2019-01-17T22:01:40 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 455,548,594 | 3,547 | abyss7 | Can't reproduce on latest master. Is it possible to provide some trivial example using console producer? | 2019-01-18T13:37:28 | 2019-01-18T13:37:28 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 455,577,260 | 4,044 | volfco | @ggservice007 You need to rewrite the data into a cluster. If you have a cluster already created, you just need to run `INSERT INTO <new distributed database> SELECT * FROM <old database>` to move the data from your single node to a cluster's distributed table. | 2019-01-18T15:08:00 | 2019-01-18T15:08:00 | {} | NONE |
ClickHouse | ClickHouse | 453,970,698 | 4,007 | filimonov | Please check the clickhouse-server logs to check what exactly was happening about that time.
It's in /var/log/clickhouse-server/ | 2019-01-14T11:13:58 | 2019-01-14T11:13:58 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 454,110,029 | 4,048 | proller | https://github.com/yandex/ClickHouse/commit/5024d1b7f762b72e7d5ca3370c4526ec204dc414
https://github.com/yandex/ClickHouse/commit/c0b72492c09164f23a68d9ad7c38dab28e6d3a63
https://github.com/yandex/ClickHouse/commit/af0b875f57aa65d5095563a218da85aa457a37b0
https://github.com/yandex/ClickHouse/commit/72df7ceee6c126086b... | 2019-01-14T18:29:03 | 2019-01-14T18:29:03 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 454,162,576 | 4,004 | alexey-milovidov | @ercolanelli-leo
`DataTypeAggregateFunction` is a full featured data type that can be stored in tables, manipulated in various ways. In contrast, `DataTypeFunction` is used only to represent temporary value for lambda expression.
You see that adding dependancy on `Context` for `AggregateFunction` is difficult...... | 2019-01-14T21:10:39 | 2019-01-14T21:10:39 | {} | MEMBER |
ClickHouse | ClickHouse | 454,242,577 | 2,586 | zhang2014 | Implementation in #4052 #4054 | 2019-01-15T02:16:41 | 2019-01-15T02:16:41 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 454,310,331 | 4,004 | filimonov | Still have doubts if such a change introduce something. If we will allow to parametrize max_itertations in a range (1 million...1 billion) some users will still have an impression that this limit is synthetic (and that's true) and they will be sure that 'for their one particular case' it should be extended to N+1 bil... | 2019-01-15T08:43:04 | 2019-01-15T08:43:04 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 455,437,807 | 4,083 | cheesedosa | So on further investigation, the issued was that I was partitioning by day. The dataset has a high cardinality along day causing increased memory usage. Changing that to the default monthly partitioning uses much more reasonable (~15GB) of memory for the insert. | 2019-01-18T06:05:59 | 2019-01-18T06:05:59 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 455,620,681 | 873 | ed-fruty | 2k19. Up ;) | 2019-01-18T17:14:50 | 2019-01-18T17:14:50 | {} | NONE |
ClickHouse | ClickHouse | 455,631,660 | 873 | blinkov | This has been partially implemented in #3946, but with some column resolution issues, so it's not announced yet. Stay tuned. | 2019-01-18T17:50:44 | 2019-01-18T17:50:44 | {
"+1": 1
} | CONTRIBUTOR |
ClickHouse | ClickHouse | 453,952,966 | 2,555 | filimonov | While trying to reproduce that on old version I've collected a lot of different memory-related segfaults. (I can attach them if you need). And also sometimes it reports it want to allocate some huge amounts of memory (like "Code: 241. DB::Exception: Received from localhost:9000, ::1. DB::Exception: Memory limit (for qu... | 2019-01-14T10:06:14 | 2019-01-14T10:07:40 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 454,817,743 | 4,004 | alexey-milovidov | Let's merge this PR first. | 2019-01-16T15:21:35 | 2019-01-16T15:21:35 | {} | MEMBER |
ClickHouse | ClickHouse | 453,926,602 | 3,976 | alesapin | Fixes https://github.com/yandex/ClickHouse/pull/4042 https://github.com/yandex/ClickHouse/pull/4041 | 2019-01-14T08:15:17 | 2019-01-14T08:15:17 | {
"+1": 1
} | MEMBER |
ClickHouse | ClickHouse | 454,168,541 | 4,052 | alexey-milovidov | Is it possible to implement automatic parameter choice based on data type? | 2019-01-14T21:29:46 | 2019-01-14T21:29:46 | {} | MEMBER |
ClickHouse | ClickHouse | 454,286,978 | 4,004 | LeoErcolanelli | @alexey-milovidov Thanks a lot for the detailed reply!
Regarding the upper bound, I understand from your message that 1 billion would be a suitable upper bound. I'll be adding a check for this upper bound in my PR.
Regarding the algorithm itself. The fact that the backtracking is sub-optimal is also something tha... | 2019-01-15T06:59:08 | 2019-01-15T06:59:08 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 454,318,694 | 4,004 | LeoErcolanelli | If we want to pass the Context down to the aggregating function I actually have a commit ready. It's a rather big change and I am not sure if I went with the best solution. If you wish I can push the commit on this branch for you to see. | 2019-01-15T09:11:28 | 2019-01-15T09:11:28 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 455,211,478 | 4,063 | alexey-milovidov | I've fixed build but your test is not passed:
```
Field nestedone_nestednestedone_nestednestednumber doesn't exist in schema CapnProto
``` | 2019-01-17T15:25:31 | 2019-01-17T15:25:31 | {} | MEMBER |
ClickHouse | ClickHouse | 455,560,606 | 4,074 | healiseu | @alexey-milovidov and @KochetovNicolai thank you for the briefing, dictionary encoded processing is an extremely important issue in database technology. I foresee that there is going to be significant progress on techniques based on this. It's all about the power of abstraction and representation that is deeply roote... | 2019-01-18T14:18:12 | 2019-01-18T18:30:36 | {} | NONE |
ClickHouse | ClickHouse | 453,969,515 | 4,051 | filimonov | Binary was compiled for a platform which not match the actual platform?
Please give more information about how did you get the binary, and what a platform you run. | 2019-01-14T11:09:15 | 2019-01-14T11:09:15 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 453,970,938 | 3,852 | alexey-milovidov | The possibility to specify `skip_nulls` for `ReplacingMergeTree` looks fine from user perspective (even if cosidering also `skip_empty_arrays` option) except that current implementation always replaces a row as a whole, not single values. Probably the implementation will be more complicated.
The possibility to spec... | 2019-01-14T11:14:50 | 2019-01-14T11:14:50 | {} | MEMBER |
ClickHouse | ClickHouse | 453,986,519 | 4,047 | filimonov | When new version will be ready it will be released. Stay tuned :)
For now, you can use 18.14.19, or build latest with needed fixes by your own. | 2019-01-14T12:19:58 | 2019-01-14T12:19:58 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 453,989,780 | 4,046 | eagle1981 | It uses default, port 8123. With no compression, if it matters. | 2019-01-14T12:32:42 | 2019-01-14T12:32:42 | {} | NONE |
ClickHouse | ClickHouse | 453,995,843 | 4,046 | filimonov | Are you 200% sure? May be you mixed ports in config (port vs http_port) or in docker port mappings?
May be only SOME of connections go to 9000?
P.S. Normally if connection goes via 8123 (http port) in stacktrace you see HTTPHandler not TCPHandler. | 2019-01-14T12:57:13 | 2019-01-14T12:57:41 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 448,435,885 | 2,719 | soheiokamoto | I've encountered exact same bug.
I was making GET request to the clickhouse server like 'https://clickhouse-hostname?query=SELECT...', and it failed with the same permission error when the whole 'query=SELECT...' is length 4077.
It works fine with other lengths shorter or longer.
Also, I was making request with my ... | 2018-12-19T01:29:15 | 2019-01-15T05:17:33 | {} | NONE |
ClickHouse | ClickHouse | 454,345,238 | 4,062 | ztlpn | This is as expected. The server will still have to look at each part to determine if the mutation changes the part or not, but the parts that are not changed will be cloned from the old parts using hard links, which should be very cheap. | 2019-01-15T10:36:14 | 2019-01-15T10:36:14 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 454,355,211 | 4,062 | vasyaabr | Thanks | 2019-01-15T11:09:02 | 2019-01-15T11:09:02 | {} | NONE |
ClickHouse | ClickHouse | 454,399,097 | 4,055 | alexey-milovidov | > Can use with raw ptr argument (which was used in Aggregator)
ColumnPtr is intrusive: you can always obtain ColumnPtr from raw ptr. | 2019-01-15T13:52:57 | 2019-01-15T13:52:57 | {} | MEMBER |
ClickHouse | ClickHouse | 454,774,824 | 4,063 | alexey-milovidov | Missing test. | 2019-01-16T13:15:51 | 2019-01-16T13:15:51 | {} | MEMBER |
ClickHouse | ClickHouse | 455,070,422 | 4,070 | justlooks | > > where I can get the new release of Tungsten replicator? should I need by it?
>
> From their "download" page it looks like it's the commercial software, not open source.
yes, it seems not fit for me :( | 2019-01-17T07:32:31 | 2019-01-17T07:32:31 | {} | NONE |
ClickHouse | ClickHouse | 455,297,546 | 4,055 | alexey-milovidov | Your own test has failed. | 2019-01-17T19:20:06 | 2019-01-17T19:20:06 | {} | MEMBER |
ClickHouse | ClickHouse | 455,456,973 | 4,070 | justlooks | > I'm not aware of any alternative similar tools. You could start with some more simple table and gradually come closer to real one to figure out what exactly is not working. Or just manually convert your data to one of formats supported by ClickHouse https://clickhouse.yandex/docs/en/interfaces/formats/
>
> Link to... | 2019-01-18T07:51:01 | 2019-01-18T07:51:01 | {} | NONE |
ClickHouse | ClickHouse | 453,943,255 | 4,046 | eagle1981 | >>clickhouse-server running on docker, tag:latest
18.16 | 2019-01-14T09:28:56 | 2019-01-14T09:28:56 | {} | NONE |
ClickHouse | ClickHouse | 454,165,124 | 4,004 | alexey-milovidov | This PR looks almost Ok, but a few considerations exist:
1. We need to also set a hard limit on number of iterations to avoid (almost) infinite loops, that even cannot be cancelled.
2. Regexp matching algorithm in `sequenceMatch` is suboptimal. The number of iterations usually grow exponentially on regexp complex... | 2019-01-14T21:18:52 | 2019-01-14T21:18:52 | {} | MEMBER |
ClickHouse | ClickHouse | 454,267,697 | 4,050 | M1ha-Shvn | Firstly, I used MergeTree engine with standard toYYYYMM('date') partitioning. It worked.
Than I changed MergeTree to ReplicatedReplacingMergeTree with (`app_id`, `toYYYYMM('date')` ) partitioning. It's more effective for me in SELECT operations. But number of parts grew fast and I decided to make partitioning back: st... | 2019-01-15T04:59:47 | 2019-01-15T05:00:40 | {} | NONE |
ClickHouse | ClickHouse | 454,348,844 | 4,026 | wawanawna | > Try to calculate MV countState or sum(1) to locate the issue (missing rows or uniqExact)
Thx for suggestion, I've tried countState and got this
```
│ 13629 │ 78100 │ 78100 │ 78100 │ 1547548020000 │
│ 13629 │ 78014 │ 78014 │ ... | 2019-01-15T10:47:53 | 2019-01-15T10:47:53 | {} | NONE |
ClickHouse | ClickHouse | 454,514,096 | 4,061 | den-crane | check privs on
ls -lta ~/.clickhouse-client-history
check that version of clickhouse-client is the same as server
try disable_suggestion -A [ --disable_suggestion ]
clickhouse-client -A
or
clickhouse-client --suggestion_limit 100 | 2019-01-15T19:12:51 | 2019-01-15T19:12:51 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 454,595,610 | 2,719 | soheiokamoto | Tried to trace the problem as best as I can. Here's the trace that I believe came from the problematic query:
```
2019.01.15 02:56:23.865627 [ 274 ] {} <Error> HTTPHandler: Code: 195, e.displayText() = DB::Exception: User default is not allowed to connect from address 2400:cb00:36:1032::a29e:4465, e.what() = DB::Exce... | 2019-01-15T23:47:06 | 2019-01-15T23:47:06 | {} | NONE |
ClickHouse | ClickHouse | 454,707,194 | 3,658 | mpolano | I also have a similar issue joining two MergeTree tables:
`SELECT distinct ip FROM measures ALL INNER JOIN netunits USING ip `
The result returns a variable number of rows at every run with a large number of duplicates.
If I swap the order of the two tables the result is correct:
`SELECT distinct ip FROM netu... | 2019-01-16T09:16:31 | 2019-01-16T16:21:33 | {} | NONE |
ClickHouse | ClickHouse | 455,043,778 | 4,079 | den-crane | What is the problem?
Mergetree read sees data just right after insert (block) end.
Do you use buffer tables? | 2019-01-17T05:05:05 | 2019-01-17T05:05:05 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 453,988,503 | 4,049 | filimonov | Wat do you see in `replicas` table?
`SELECT * FROM system.replicas`
| 2019-01-14T12:27:49 | 2019-01-14T12:27:49 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 454,000,038 | 4,049 | crrlcx | From first node:
```shell
:) SELECT * FROM system.replicas WHERE table='tbl_test' FORMAT Vertical
SELECT *
FROM system.replicas
WHERE table = 'tbl_test'
FORMAT Vertical
Row 1:
──────
database: chdata
table: tbl_test
engine: ReplicatedReplacingMer... | 2019-01-14T13:13:12 | 2019-01-14T13:13:12 | {} | NONE |
ClickHouse | ClickHouse | 454,349,525 | 4,004 | filimonov | > If we want to pass the Context down to the aggregating function I actually have a commit ready.
No. As @alexey-milovidov said it's not the best idea.
But beside user level config, there is also a server config - config.xml (from poco config() ).
But I think it's worth to try to improve the performance first,... | 2019-01-15T10:50:14 | 2019-01-15T11:38:40 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 454,391,772 | 4,058 | ztlpn | Спасибо за репорт, удалось повторить сходным скриптом. Разбираюсь. | 2019-01-15T13:28:46 | 2019-01-15T13:28:46 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 455,069,436 | 4,070 | blinkov | > where I can get the new release of Tungsten replicator? should I need by it?
From their "download" page it looks like it's the commercial software, not open source. | 2019-01-17T07:27:41 | 2019-01-17T07:27:41 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 455,542,794 | 2,675 | abyss7 | First attempt is here #4094 | 2019-01-18T13:17:43 | 2019-01-18T13:17:43 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 454,687,345 | 4,049 | M1ha-Shvn | http://prntscr.com/m7tkhm | 2019-01-16T08:04:39 | 2019-01-16T08:04:39 | {} | NONE |
ClickHouse | ClickHouse | 454,813,597 | 4,015 | sjoerdmulder | Probally this regresssion is caused by #3673
But more curius is why an existing test is altered from `MATERIALIZED` to `DEFAULT` instead of a new test added.
https://github.com/yandex/ClickHouse/pull/3673/files#diff-0ea8877ed15b4be113be1a0b231dbc3aR22 | 2019-01-16T15:10:39 | 2019-01-16T15:10:39 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 454,865,446 | 4,004 | alexey-milovidov | PS. If the number of states is small (for example, 16), we can use extremely efficient implementation based on table lookup (for every combination of active states). | 2019-01-16T17:24:17 | 2019-01-16T17:24:17 | {} | MEMBER |
ClickHouse | ClickHouse | 453,966,158 | 4,047 | eagle1981 | Looks like it is. Any plans to promote changes to :latest ? | 2019-01-14T10:56:00 | 2019-01-14T10:56:00 | {} | NONE |
ClickHouse | ClickHouse | 454,063,506 | 4,004 | LeoErcolanelli | @filimonov I gave a try at passing the context down to the `AggregateFunctions` however I encounter 1 issue and I think your greater knowledge of the code base will be of help.
Changing the definition of `AggregateFunctionCreator` in order to accept an additional `Context` argument is pretty straightforward. The iss... | 2019-01-14T16:17:21 | 2019-01-14T16:17:21 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 455,068,080 | 4,070 | justlooks | > Altinity replicator works in production in some companies, so it is reliable enough. Thanks for the problem report, we will look what's wrong.
>
> Continuent recently released replication to ClickHouse as well, you can try it out: https://www.continuent.com/replicating-data-into-clickhouse/
where I can get the... | 2019-01-17T07:21:45 | 2019-01-17T07:21:45 | {} | NONE |
ClickHouse | ClickHouse | 455,137,079 | 3,960 | ztlpn | UBsan failure is an unrelated bug in the `bitTest` function:
```
$ ./clickhouse local -q 'SELECT bitTest(-607446, -1687822.6)'
../dbms/src/Functions/bitTest.cpp:15:62: runtime error: shift exponent -1687822 is negative
``` | 2019-01-17T11:20:14 | 2019-01-17T11:20:14 | {} | CONTRIBUTOR |
ClickHouse | ClickHouse | 455,145,342 | 4,079 | ggservice007 | Later, I will provide some detail info. | 2019-01-17T11:51:48 | 2019-01-17T11:51:48 | {} | NONE |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.